Abstract 17352: Incidence and Predictors of Stroke in Patients with Chronic Heart Failure: Does Left Ventricular Function Matter? Insights From the CHARM Program
Objectives Patients with heart failure and reduced ejection fraction (HF-REF) are at increased risk of stroke, with some studies suggesting an inverse association between ejection fraction (EF) and stroke risk. The rate of stroke in simultaneously enrolled patients with HF-REF and preserved EF (HF-PEF) has not been reported and the predictors of stroke in a contemporary population of patients with HF are unknown. The Candesartan in Heart Failure: Assessment of Reduction in Mortality and morbidity (CHARM) program, which included patients with a wide range of EF, provided a unique opportunity to examine these questions about stroke in HF.
Methods We evaluated risk factors for stroke with multivariable Cox proportional hazard regression analysis using forward- and backward-stepwise selection of candidate variables, with gender, history of hypertension, EF and randomized treatment (i.e. candesartan) forced into the model.
Results Of the 7599 participants in CHARM, 287 (3.8%) experienced a fatal or nonfatal stroke during a median follow-up of 37.7 months. The stroke rate was 1.3 (95% CI=1.1-1.5) and 1.4 (95% CI=1.2-1.7) per 100 pt yrs in subjects with reduced and preserved EF respectively. The 5 strongest predictors of stroke (ranked by chi-square value) were: older age, previous stroke, higher systolic blood pressure, diabetes and atrial fibrillation on baseline ECG; other independent predictors are shown in the figure. Univariate predictors not retained in the multivariable model were: history of hypertension, NYHA class III or IV, baseline oral anticoagulant use, previous coronary bypass surgery, current smoking, signs of HF, diuretic use and female sex; none of the other forced variables were predictive in either the univariate or multivariable models.
Conclusions The rate of stroke is similar in patients with HF-REF and HF-PEF. Even in this relatively normotensive population, higher systolic BP was associated with greater stroke risk.
- © 2011 by American Heart Association, Inc.